Instagram Cyberbullying Data with Topics and Severity Labels

This dataset, titled Insta-CTSR, contains annotated Instagram data related to cyberbullying, including topics, severity, and participant roles. It was created by Manuel Sandoval Madrigal and is hosted by Harvard Dataverse. The specific number of rows, columns, and file formats are not provided in the input.

Use Cases

• Classify cyberbullying severity levels using the annotated severity labels.
• Analyze the distribution of discussion topics within cyberbullying incidents.
• Identify and model the roles of participants (e.g., bully, victim, bystander) in online interactions.
• Train a multi-label model to predict both the topic and severity of a cyberbullying post.

Strengths

• Dataset is hosted on the authoritative Harvard Dataverse platform.
• Focuses on the specific and relevant social media platform Instagram.
• Includes multiple analytical dimensions: cyberbullying, topics, severity, and roles.

Limitations

• The dataset size (rows) and feature scope (columns) are unknown, limiting assessment of statistical power.
• Data collection methodology and potential biases (e.g., geographic, linguistic) are not specified.
• The description lacks sample data, preventing verification of annotation quality and format.

Provenance

Source

Harvard Dataverse

Freshness

The metadata was last updated on January 30, 2026, indicating recent maintenance.